A process capacity sheet is a standardized-work form that records, for each machine in a linked process, the manual time, the automatic machine time, and the tool-change time and interval, then calculates how many parts that machine can make per shift, so you can find the real bottleneck before you balance the line.
It is the least glamorous of the three standardized-work documents and the one that saves the most bad decisions. Balance a line by eye and you almost always protect the wrong station; a process capacity sheet replaces the eye with arithmetic, machine by machine, including the tool changes that quietly eat capacity nobody counted. This guide covers what goes on the sheet, how the capacity math actually works, and how the sheet feeds line balancing and takt.
What is a process capacity sheet?
A process capacity sheet is one of the three core forms of Toyota-style standardized work, alongside the standardized work combination table and the standardized work chart. The Lean Enterprise Institute defines standardized work as built on three elements, takt time, the work sequence, and the standard in-process stock, and documented on these shop-floor forms (Lean Enterprise Institute, Standardized Work). The capacity sheet is the one that answers a single blunt question: how many good parts can each machine in this process actually produce in the time available?
The word capacity here is specific. It is not the nameplate rating a vendor prints on a machine; it is the real, sustainable output once you subtract the time the machine spends changing tools and account for the manual work an operator must do at it. Those two subtractions are exactly what a casual estimate misses, and they are why the slowest station on paper is often not the true constraint on the floor.
What goes on a process capacity sheet?
The sheet is a row per machine and a handful of columns that build up to a capacity number. The columns matter because each one is a place capacity leaks.
| Column | What it records | Why it matters |
|---|---|---|
| Process / machine | Step number, part, and machine ID | Ties the row to a real station in sequence |
| Manual time | Operator hand time at the machine per piece | Work that cannot run in parallel with the machine |
| Auto (machine) time | Unattended machine run time per piece | Runs while the operator moves on |
| Completion time | Manual plus auto time for one piece | The base per-piece time before tool changes |
| Tool-change interval | Pieces made between tool changes | How often you pay the change penalty |
| Tool-change time | Minutes to swap the tool | Capacity lost to setup, spread over the run |
| Capacity | Good pieces per shift for this machine | The number you compare across the line |
The two tool-change columns are the sheet's whole reason for existing. A machine with a 20-second completion time looks fast until you learn it needs a five-minute tool change every 50 pieces, at which point six seconds per piece vanish into changeovers. Only a form that forces you to write the interval and the change time turns that invisible loss into a number.
How do you calculate capacity on the sheet?
The core formula divides the time you have by the time each piece really costs, including its share of tool changes. Written plainly:
Capacity per shift = available operating time / (completion time per piece + tool-change time per piece) where completion time is manual plus auto time, and tool-change time per piece is the change time divided by the number of pieces made between changes.
- Set the available operating time. Take the shift length and subtract planned stops, breaks, meetings, and scheduled maintenance. This is the real time the machine can run, in seconds.
- Measure manual and auto time. Time the operator's hand work at the machine and the machine's unattended run time, per piece, by observation, not from the spec sheet.
- Add them into completion time. Completion time per piece is manual time plus auto time, the base cost of one good part before tool changes.
- Convert the tool change to a per-piece cost. Divide the tool-change time by the number of pieces made between changes. A five-minute change every 50 pieces is six seconds per piece.
- Divide to get capacity. Divide available operating time by the sum of completion time and tool-change time per piece. The result is good pieces per shift for that machine.
- Repeat for every machine and compare. The machine with the lowest capacity is the constraint, and it sets what the whole process can actually deliver.
Worked once, the arithmetic is simple; the value is in doing it for every machine with honest, observed numbers. That is when the surprise usually lands: the station everyone blamed is fine, and a quiet machine two steps upstream, bleeding capacity to frequent tool changes, is the true constraint.
By the numbers: the three standardized-work documents
The process capacity sheet is not a standalone worksheet; it is the foundation of a set. Toyota-style standardized work is documented on three linked forms, the process capacity sheet (capacity per machine), the standardized work combination table (how manual, auto, and walk time fit within takt), and the standardized work chart (layout and standard in-process stock), and all three rest on takt time, a defined work sequence, and standard in-process stock (Lean Enterprise Institute, Standardized Work). The capacity sheet comes first because the other two need its numbers: you cannot combine work against takt until you know what each machine can actually produce. Where Harmony fits: a capacity sheet is timed by hand on one good day, but tool wear, changeover skill, and micro-stops shift the real numbers shift to shift. When manual time, machine time, and stops are captured live at each station, the capacity on the sheet reflects the line as it runs now, not as it ran the morning someone stood there with a stopwatch.
How does the capacity sheet relate to takt and line balancing?
The capacity sheet answers "what can each machine make?"; takt time answers "what must the line make to meet demand?"; and line balancing is the act of reconciling the two. You cannot balance a line honestly without the capacity sheet, because balancing means moving work off the constraint and onto machines with spare capacity, and only the sheet tells you which machines those are. Compare each machine's capacity to the takt requirement, and any machine below it is a constraint you must break, by cutting its tool-change loss, offloading manual work, or reducing its cycle, before the line can hold rate.
This is also where the capacity sheet meets flow. A value stream map shows where material waits across the whole plant; the capacity sheet zooms into one process and explains why a step cannot keep up, in terms of manual time, machine time, and tool changes you can actually attack. The two work together: the map finds the slow process, the capacity sheet finds the slow machine inside it.
What are the most common capacity sheet mistakes?
The first is using nameplate or spec numbers instead of observed ones. A machine's rated cycle is a marketing figure; its real completion time includes the operator's hand work and the small hesitations that never make the brochure. The second is ignoring tool changes, which is the exact loss the sheet exists to capture, and the most common way a bottleneck stays hidden.
The third is treating the sheet as a one-time study. Tool life shifts as materials and speeds change, operators get faster or slower, and a machine that had spare capacity in January is the constraint by June. A capacity sheet timed once and filed is a snapshot of a line that no longer exists. The fourth is confusing capacity with output: capacity is what a machine can make under standard conditions, while actual output also loses time to unplanned downtime which belongs in a separate reckoning.
How does a process capacity sheet connect to the floor?
The capacity sheet is the arithmetic backbone of standard work and it feeds directly into line balancing takt planning, and constraint management. It also pairs naturally with a product-quantity analysis: once you know which products run in volume, you build capacity sheets for the machines those high-runners depend on, and design flow around the real constraints. All of it lives inside lean manufacturing where the whole point is to make the line produce to takt without heroics.
The hard part is keeping the numbers true. A capacity sheet is only as good as its inputs, and its inputs, manual time, machine time, tool-change frequency, all drift the moment you stop watching. Timed by hand, they are accurate for a day and stale by the next changeover. When manual time, machine cycle, and every stop are captured live at each station, the capacity sheet can be recomputed against reality, and a machine sliding toward the constraint shows up before it starves the line. That live picture is what Harmony gives a plant through station-level capture and it is the difference between a capacity sheet that guides balancing and one that misleads it. CLS made exactly that shift, from cycle and stop data found the next morning to data visible during the shift. No rip-and-replace.